Logical Expression Evaluation on ListOps-O Length Generalization (Lengths 200-300)
99.9AccuracyEBT-GRC
Evaluation Results
| Method | Links | |
|---|---|---|
| EBT-GRC2023.11 | 99.9 | |
| OM2023.11 | 99.6 | |
| CRVNN2023.11 | 99.5 | |
| BT-GRC OS2023.11 | 99.5 | |
| RIR-EBT-GRC2023.11 | 99.15 | |
| RIR-EBT-GRC (-S4D)Backbone=S4D2023.11 | 99.15 | |
| RIR-EBT-GRC (-Beam Align)Alignment=Beam Align2023.11 | 91.75 | |
| MEGA2023.11 | 45.21 | |
| BBT GRC2023.11 | 43.6 | |
| RIR-GRC2023.11 | 41.75 | |
| S4D2023.11 | 31 |